In the highly heterogeneous central nervous system, the genes of greatest interest are those involved in the unique functions of each neuronal phenotype, and with their dysfunctions related to disease, aging, and injury. The objective of this project is to implement a generalizable stategy to identify these genes. The first aim is to develop methods to isolate phenotypically similar neurons and specifically identify cDNAs from mRNAs that are critical to phenotype and function. Transgenic mice expressing degradable green fluorescent protein (dGFP) from the per1 promoter, a clock-element gene, or from the fos promoter, a clock-driven gene, allow automated isolation of suprachiasmatic nucleus neurons expressing a circadian rhythm. Developing methods to use antisense RNA amplification, a linear mechanism of cDNA amplification from single neurons, and representational difference analysis, a high fidelity method of differential cDNA subtraction, on single phenotypically sorted neurons will allow both broad characterization of the mRNAs they express and specific identification of mRNAs whose expression changes during the circadian rhythm. Rather than assuming that changes in mRNA expression necessarily lead to protein changes, the second aim will reduce to practice a novel method for improved sensitivity in Western blots and use it to test rhythms in protein expression. The third aim seeks to develop a method to simultaneously measure multiple proteins from single neurons, using a high-throughput optical method that is predicted to detect antigens at levels sufficient for analysis of proteins in single neurons.